Saccharomyces Boulardii for Frequent Diarrhea

Saccharomyces boulardii is not a bacterium. It is a yeast, and that single fact explains most of its clinical appeal: antibiotics cannot kill it, so it can be taken at the same time as the drug that is causing the diarrhoea in the first place. Meta-analyses consistently show it reduces the incidence of antibiotic-associated diarrhoea in both adults and children, and it has supporting evidence in travellers' diarrhoea, acute infectious diarrhoea and as an adjunct in Clostridioides difficile prevention. Effect sizes are moderate, and timing is what determines whether you see them. Where it does not belong is chronic, unexplained diarrhoea. Symptoms lasting beyond two to four weeks need a diagnosis, because coeliac disease, inflammatory bowel disease, bile acid diarrhoea and microscopic colitis all look like a sensitive gut from the outside.

The strain identity on the label matters. Almost all trial evidence uses S. boulardii CNCM I-745, the preparation sold in Europe as a licensed medicine in some countries. Generic products labelled only S. cerevisiae var. boulardii may or may not behave the same way. Dose in colony-forming units also matters less than consistency: 5 to 10 billion CFU daily, started on day one of the antibiotic and continued for a week or two afterwards, is the pattern that worked.

S. boulardii does several things a bacterial probiotic cannot. It secretes a 54 kDa protease that cleaves C. difficile toxin A and its intestinal receptor, and a phosphatase that dephosphorylates E. coli endotoxin, neutralising two of the direct drivers of secretory diarrhoea. It also acts structurally and immunologically. It binds pathogens to its own cell surface and carries them out with transit, preserves tight junction integrity, stimulates secretory IgA and brush border enzyme expression including disaccharidases, and supports short-chain fatty acid production by the residual microbiota. Because it is a yeast, it is intrinsically resistant to antibacterial drugs, does not transfer antibiotic resistance genes, and clears from the gut within days of stopping.

The regimen used in positive trials is 250 to 500 mg twice daily, equivalent to roughly 5 to 10 billion CFU per day. Unlike bacterial probiotics, doses do not need to be separated from antibiotics, which removes the main reason people get the timing wrong. Start on the first day of the antibiotic rather than waiting for symptoms, and continue for one to two weeks after the course ends while the resident microbiota re-establishes. Capsules can be opened into cool food or drink, but not anything hot, which kills the yeast.

A 2015 systematic review with meta-analysis found S. boulardii significantly reduced the risk of antibiotic-associated diarrhoea, and a 2010 meta-analysis in adults reached the same conclusion across a broader set of diarrhoeal indications. A 2017 meta-analysis of probiotics in outpatients supports the class effect in the setting where most antibiotics are prescribed. The limits are worth stating plainly. Baseline diarrhoea rates in trial populations varied widely, blinding was inconsistent in older studies, and pooled estimates are driven by relative risk reductions applied to modest absolute risks. Benefit is clearest when baseline risk is high: broad-spectrum antibiotics, older adults, hospital settings and previous episodes.

In immunocompetent people, S. boulardii is well tolerated. Reported effects are bloating, flatulence, constipation and occasional thirst, usually mild and self-limiting. The serious risk is fungaemia, and it is not theoretical. Cases have occurred in critically ill patients, people with central venous catheters, those on broad immunosuppression and premature infants, sometimes from environmental contamination of the line rather than gut translocation. It is one of the few probiotics with a documented invasive infection signal, which is why intensive care and haematology settings usually prohibit it.

There are very few pharmacokinetic interactions. The clinically relevant conflicts are with antifungals, which kill the organism, and with situations where a live yeast should not be introduced at all.